app.py

import gradio as gr
import numpy as np
import random
import torch
from diffusers import DiffusionPipeline
import os

# Constants
MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = 2048
DEFAULT_IMAGE_SIZE = 1024

# Model setup
dtype = torch.bfloat16
device = "cuda" if torch.cuda.is_available() else "cpu"
huggingface_token = os.getenv("HUGGINGFACE_TOKEN")

pipe = DiffusionPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-dev",
    torch_dtype=dtype,
    token=huggingface_token
).to(device)

def generate_image(prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, progress=gr.Progress(track_tqdm=True)):
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    generator = torch.Generator().manual_seed(42)
    image = pipe(
        prompt=prompt,
        width=width,
        height=height,
        num_inference_steps=num_inference_steps,
        generator=generator,
        guidance_scale=guidance_scale
    ).images[0]
    return image, seed

css = """
body {
    background-color: #f4faff;
    color: #005662;
    font-family: 'Poppins', sans-serif;
}
.container {
    margin: 0 auto;
    max-width: 900px;
    padding: 20px;
}
.gr-button {
    background-color: #0288d1;
    color: white;
    border-radius: 8px;
    transition: background-color 0.3s ease;
}
.gr-button:hover {
    background-color: #0277bd;
}
.gr-box {
    border-radius: 12px;
    border: 1px solid #eeeeee;
}
"""

with gr.Blocks(css=css, theme=gr.themes.Soft(primary_hue="blue", secondary_hue="gray")) as demo:
    gr.Markdown("""
    # FLUX.1 [dev] | A Text-To-Image Rectified Flow 12B Transformer
    
    Enter a text prompt below to generate an image. Click 'Generate' to create your image.
    """)
    
    with gr.Row():
        with gr.Column(scale=4):
            prompt = gr.Text(
                label="Prompt",
                placeholder="Enter your prompt here",
                lines=3
            )
        with gr.Column(scale=1):
            generate_button = gr.Button("Generate", variant="primary")
    
    result = gr.Image(label="Generated Image", type="pil")
    
    with gr.Accordion("Advanced Settings", open=False):
        seed = gr.Slider(label="Seed", minimum=0, maximum=MAX_SEED, step=1, value=0)
        randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
        
        with gr.Row():
            width = gr.Slider(label="Width", minimum=256, maximum=MAX_IMAGE_SIZE, step=32, value=DEFAULT_IMAGE_SIZE)
            height = gr.Slider(label="Height", minimum=256, maximum=MAX_IMAGE_SIZE, step=32, value=DEFAULT_IMAGE_SIZE)
        
        with gr.Row():
            guidance_scale = gr.Slider(label="Guidance Scale", minimum=1, maximum=15, step=0.1, value=3.5)
            num_inference_steps = gr.Slider(label="Number of inference steps", minimum=1, maximum=50, step=1, value=28)
    
    gr.Examples(
        examples=[
            "a galaxy swirling with vibrant blue and purple hues",
            "a futuristic cityscape under a dark sky",
            "a serene forest with a magical glowing tree",
            "a portrait of a smiling woman with a colorful floral crown",
            "a fantastical creature with the body of a dragon and the wings of a butterfly",
        ],
        inputs=prompt,
        outputs=[result, seed],
        fn=generate_image,
        cache_examples=True,
    )
    
    generate_button.click(
        fn=generate_image,
        inputs=[prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
        outputs=[result, seed]
    )


demo.launch(share=True)